This proposal focuses on the use of molecular dynamics (MD) simulations to study dihydrofolate reductases. Two specific areas will be addressed: 1) elucidating the relationship between ligand binding and the preferred conformation of the M20 loop of chromosomal E. coli dihydrofolate reductase (DHFR) and 2) determining the source of binding cooperativity in an R-plasmid-encoded R67 DHFR. These investigations will lead to a better overall understanding of how enzymes function. This information will be useful for designing enzymes and for designing new inhibitors of these particular enzymes. For project one, a series of molecular dynamics simulations will be performed on different complexes and conformations. MM/PBSA calculations will then be used to obtain relative free energies. The results will indicate how the M20 loop potential energy surface is affected by the presence of substrate or co-factor and what the preferred M20 loop conformation is for each complex. In addition, the calculations on the catalytic complexes and on the model complexes used experimentally will be compared. This data will be used to assess the ability of the model experimental complexes to accurately represent the catalytic ones. Finally, a theoretical model of how the M20 loop conformation changes during the catalytic cycle will be developed. For project two, the binding cooperativity of R67 DHFR will be studied by performing molecular dynamics simulations on several complexes and mutants. Examining changes in enzyme structure and dynamical behavior between the apo-enzyme, binary complexes and ternary complexes will indicate if the protein plays a role in binding cooperativity. In addition, a detailed examination of the contacts between the ligand and protein and between the ligands will help determine why FOL or DHF binding is positively cooperative but NADPH binding is negatively cooperative. The results will also provide information on how identical binding sites can be used to bind both NADPH and DHF. [unreadable] [unreadable]